Update app.py

app.py

@@ -1,24 +1,51 @@
-import pandas as pd
-import tabula
+import re
+from typing import Dict, List, Union
+import gradio as gr
 
-def extract_data(pdf_file):
-    # Extract data from the PDF file using tabula
-    tables = tabula.read_pdf(pdf_file, pages='all')
+def parse_federal_transformers(file_text: str) -> Dict[str, Union[str, List[Dict[str, Union[str, int, float]]]]]:
+    parsed_data = {
+        "Purchase Order No": "",
+        "Date": "",
+        "Supplier": "Federal Transformers Co. LLC",
+        "Invoice Address": "",
+        "Delivery Address": "",
+        "Currency": "",
+        "Payment Terms": "",
+        "Items": []
+    }
 
-    # Combine the extracted tables into a single DataFrame
-    data = pd.concat(tables, ignore_index=True)
+    try:
+        # Extract Purchase Order No and Date
+        parsed_data["Purchase Order No"] = re.search(r"Purchase Order No\.\s(\d+)", file_text).group(1)
+        parsed_data["Date"] = re.search(r"Date:\s+(\d{2}-\w{3}-\d{2})", file_text).group(1)
 
-    # Rename columns to match the expected output format
-    data.columns = ['Purchase Order No', 'Date', 'Material Description', 'Unit', 'Quantity', 'Dely Qty', 'Dely Date', 'Unit Rate', 'Value']
+        # Extract Invoice Address and Delivery Address
+        parsed_data["Invoice Address"] = re.search(r"Invoice Address\s*:\s*(.*?)(?=\sDelivery Address)", file_text, re.DOTALL).group(1).strip()
+        parsed_data["Delivery Address"] = re.search(r"Delivery Address\s*:\s*(.*?)(?=\sNote)", file_text, re.DOTALL).group(1).strip()
 
-    # Remove any unnecessary rows and columns
-    data = data.dropna(how='all')
+        # Define pattern for extracting item details
+        item_pattern = re.compile(r"(\d+)\s+([\w\s]+)\s+(\d+)\s+([\d.]+)\s+([\d.]+)")
+        for match in item_pattern.finditer(file_text):
+            parsed_data["Items"].append({
+                "Item No": match.group(1),
+                "Description": match.group(2).strip(),
+                "Quantity": int(match.group(3)),
+                "Unit Price": float(match.group(4)),
+                "Total Price": float(match.group(5)),
+            })
+    except Exception as e:
+        print(f"Error parsing Federal Transformers PO: {e}")
+    
+    return parsed_data
 
-    return data
+# Create Gradio interface
+iface = gr.Interface(
+    fn=parse_federal_transformers,
+    inputs="text",
+    outputs="json",
+    title="Federal Transformers PO Parser",
+    description="Upload the text of a Federal Transformers purchase order to extract details."
+)
 
 if __name__ == "__main__":
-    pdf_file = 'your_pdf_file.pdf'
-    data = extract_data(pdf_file)
-
-    # Save the extracted data to an Excel file
-    data.to_excel('output.xlsx', index=False)
+    iface.launch()